Figure - PMC

Skip to main content

An official website of the United States government

Here's how you know

Here's how you know

Official websites use .gov
A .gov website belongs to an official government organization in the United States.

Secure .gov websites use HTTPS
A lock ( ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites.

View full-text article in PMC

. 2022 Oct 31;19(11):1427–1437. doi: 10.1038/s41592-022-01652-7

Search in PMC
Search in PubMed
View in NLM Catalog
Add to search

© This is a U.S. Government work and not under copyright protection in the US; foreign copyright protection may apply 2022

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.

PMC Copyright notice

Extended Data Fig. 4 — a) 3D rendering of nuclei expressed in live C. elegans embryo collected by diSPIM and processed with single-input RLN and dual-input RLN. b) Higher magnification of white rectangle in a), comparing raw views A/B, joint deconvolution of views A/B (GT), single- and dual-input RLN predictions. The dual-input RLN uses the information available in both views to achieve resolution isotropy, providing near-identical output to the ground truth. c) 3D rendering of membrane marker in live C. elegans embryo, raw data collected by diSPIM and processed with dual-input RLN. d) Higher magnification of white rectangle in c), comparing raw views A/B joint deconvolution (GT), and dual-view RLN predictions. Dual-input RLN uses complementary dual-view information to enable near-isotropic resolution, providing results near-identical to the ground truth (magenta and green arrows). e) Higher magnification of dashed yellow and magenta parallelograms in c), comparing dual-view deconvolution ground truth, single- and dual-view RLN predictions. Dual-input RLN better recovers axial resolution than single-input RLN. f) SSIM analysis for data displayed in b), e), showing that RLN-dual better recovers the signals than RLN-single when the views are contaminated by scattering. Individual values (open circles), means, and standard deviations from N = 9 volumes are shown for membranes and nuclei. g) Maximum-intensity projections of raw images of C. elegans embryo expressing GCaMP3 acquired with reflective diSPIM, deconvolution based on spatially invariant PSF, deconvolution based on spatially variant PSF (ground truth ‘GT’), single-input RLN, and dual-input RLN predictions (using two orthogonal views as input). Raw input is contaminated with spatially varying epifluorescence background (red arrows). Spatially variant deconvolution, single-input RLN, and dual-input RLN remove associated epifluorescence contamination, enhancing resolution and contrast. PSNR and SSIM analysis confirms this result (Raw: PSNR 22.65 ± 1.46, SSIM 0.52 ± 0.04; spatially invariant deconvolution: PSNR 26.61 ± 1.53, SSIM 0.62 ± 0.02; single-input RLN: PSNR 31.56 ± 0.70, SSIM 0.84 ± 0.02; dual-input RLN: PSNR 34.55 ± 0.61, SSIM 0.87 ± 0.01; N = 12 volumes). Scale bars: a, b, d, e) 5 µm, c, g) 10 µm.